Enhancing Document Retrieval in COVID-19 Research: Leveraging Large Language Models for Hidden Relation Extraction

2025-06-23

AI Summary

The surge in COVID-19 literature has exposed limitations of conventional retrieval systems, which rely on shallow syntactic parsing tools and struggle to identify latent semantic relationships in unlabeled text, thereby constraining knowledge discovery. To address this, we propose the first LLM-driven implicit relation extraction method specifically designed for COVID-19 literature. Our approach deeply integrates large language models into the retrieval pipeline and couples them with the Covrelex-SE framework to build a semantics-enhanced retrieval system. Crucially, it operates without manual annotation, automatically uncovering deep entity-level associations—such as “drug–target–pathway”—beyond surface-level syntactic patterns. Experimental results demonstrate significant improvements in both precision and recall, markedly enhancing the relevance and conceptual depth of retrieved documents. This advancement delivers more accurate, insightful, and knowledge-rich literature support to researchers.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Natural Language Processing: Information Extraction
• Computer Vision: Image and Video Retrieval

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Large language models for search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

In recent years, with the appearance of the COVID-19 pandemic, numerous publications relevant to this disease have been issued. Because of the massive volume of publications, an efficient retrieval system is necessary to provide researchers with useful information if an unexpected pandemic happens so suddenly, like COVID-19. In this work, we present a method to help the retrieval system, the Covrelex-SE system, to provide more high-quality search results. We exploited the power of the large language models (LLMs) to extract the hidden relationships inside the unlabeled publication that cannot be found by the current parsing tools that the system is using. Since then, help the system to have more useful information during retrieval progress.

Problem

• Enhancing COVID-19 document retrieval efficiency
• Extracting hidden relations using large language models
• Improving search result quality in pandemic research

Innovation

• Leveraging LLMs for hidden relation extraction
• Enhancing retrieval with unlabeled publication analysis
• Improving search results quality via advanced parsing